This week highlighted the growing importance of transparent AI operations and persistent context in creating trustworthy user experiences.
As AI becomes more capable and autonomous, user trust depends increasingly on transparent operations and persistent context rather than just performance metrics.
This Week in AI Products
 Cursor |
Mar 21 |
Cursor admits to using China's Kimi model in Composer 2 after user backlash
Cursor faced criticism for not initially disclosing that their Composer 2 coding model was built using China's Kimi K2.5 model. The company later admitted the dependency after developers questioned the model's origins and performance characteristics. This transparency issue highlights growing concerns about model provenance in AI development tools. Source →
Designer's Takeaway: Design clear model attribution and data source disclosure patterns into your AI features from day one. Users increasingly expect transparency about which models power their tools, especially when working with sensitive code or data.
Pattern: Responsible AI Design
 Claude |
Mar 22 |
Anthropic introduces Code Channels for persistent AI agent workflows
Anthropic launched Code Channels, a new feature that enables persistent, collaborative AI workflows similar to OpenAI's canvas-style interfaces. The feature allows developers to maintain ongoing conversations and code sessions with Claude, creating a more continuous development experience that maintains context across sessions. Source →
Designer's Takeaway: Design for persistent context and memory in AI interactions rather than treating each session as isolated. Consider how maintaining conversation history and project context can significantly improve user experience in professional workflows.
Pattern: Selective Memory
 WordPress |
Mar 20 |
WordPress.com enables direct AI agent actions through natural conversation
WordPress.com announced that AI agents like Claude, ChatGPT, and Cursor can now take direct actions on websites through conversational interfaces. Users can ask AI agents to publish posts, modify content, or manage site settings using natural language commands. This integration removes traditional UI barriers between AI assistance and content management workflows. Source →
Designer's Takeaway: Consider how conversational interfaces can replace traditional form-based workflows while providing clear preview and confirmation patterns. Design natural language interactions that feel intuitive while showing users exactly what actions the AI will take on their behalf.
Pattern: Conversational UI
 Microsoft Copilot |
Mar 20 |
Microsoft reduces AI integrations across Windows apps
Microsoft is scaling back Copilot integrations from several Windows applications including Photos, Widgets, Notepad, and other native apps. This move appears to address user complaints about AI feature bloat and overly aggressive integration of AI assistance in everyday workflows. Source →
Designer's Takeaway: Apply restraint when integrating AI features into existing products. Even Microsoft recognizes that users can feel overwhelmed by too many AI entry points. Focus on selective and meaningful AI integration rather than comprehensive coverage of every possible touchpoint.
Pattern: Contextual Assistance
 Perplexity |
Mar 20 |
Partners with b.well for AI-powered medical record search
Perplexity partnered with b.well Connected Health to enable AI-powered search through personal medical records and health data. The integration allows users to ask natural language questions about their medical history, treatment plans, and health insights while maintaining data privacy and security standards. Source →
Designer's Takeaway: Consider how AI can make complex, scattered data more accessible through natural language interfaces. Design for sensitive data contexts by emphasizing privacy safeguards and transparent data handling throughout your interface communications.
Pattern: Conversational UI
 Notion |
Mar 20 |
AI Meeting Notes can now run in the background
Notion updated its AI Meeting Notes feature to run in the background, allowing users to capture meeting insights without actively managing the AI tool during calls. This reduces cognitive overhead while maintaining comprehensive documentation of meeting content. Source →
Designer's Takeaway: Design AI features to work invisibly in the background when possible, reducing user cognitive load while still delivering value. Provide clear but unobtrusive status indicators so users know the AI is working without requiring constant attention or management.
Pattern: Ambient Intelligence
 OpenAI |
Mar 19 |
Chain-of-thought monitoring detects coding agent misalignment
OpenAI revealed how they monitor internal coding agents for potential misalignment using chain-of-thought analysis. The approach examines the reasoning process of AI agents working on code to identify when they might be acting contrary to intended goals or safety guidelines. Source →
Designer's Takeaway: Consider exposing AI reasoning processes to users when the stakes are high. Design interfaces that show not just what the AI is doing, but why it's making specific choices, especially for sensitive tasks like code generation or data analysis.
Pattern: Explainable AI (XAI)
| LinkedIn |
Mar 20 |
Bans AI agent that successfully operated as virtual cofounder
LinkedIn banned an AI agent that had been successfully operating as a virtual cofounder, participating in professional conversations and even receiving corporate speaking invitations. The incident highlights the tension between platform policies and AI agent capabilities as these tools become increasingly sophisticated at mimicking human behavior. Source →
Designer's Takeaway: Design clear disclosure mechanisms when AI agents interact in human-centric platforms. Consider the ethical implications of AI agents that can convincingly impersonate humans and build transparency indicators into agent interfaces to maintain trust and comply with platform policies.
Pattern: Responsible AI Design
Steal This Week
Claude's Code Channels for persistent workflows
This feature transforms AI from a series of isolated interactions into a continuous collaborative partner. By maintaining context across sessions, users can build incrementally on previous work, creating a much more natural and productive workflow than starting fresh each time.
Pattern to Know
Responsible AI Design
Multiple major incidents this week highlighted the critical importance of transparency and ethical AI deployment. From Cursor's model disclosure controversy to LinkedIn's AI agent ban, users and platforms are demanding clearer boundaries and honest communication about AI capabilities and origins.
When to use it: Apply this pattern whenever AI systems interact with sensitive data, make autonomous decisions, or could be mistaken for human behavior. Essential for maintaining user trust and regulatory compliance.
Want the full breakdown on any pattern mentioned above?
Explore All 28 Patterns →